Antibodies can be made to function as enzymes bec ause they possess a stable structural framework that can support a number of different combining site mutations. This allows for sampling of a wide variety of chemical environments, including those that are amenable to chemical catalysis. Nature has selected the alpha/beta-barrel fold as a similar adaptable framework for the divergent evolution of new enzymes. Approximately 10% of all known enzymes are members of the alpha/beta barrel family and their enzymatic activities, catalytic efficiencies, and cofactor usage vary widely. Phosphotriesterase (PTE) is an a/b barrel enzyme from soil bacteria which catalyzes the hydrolysis and detoxification of phosphotriester insecticides and nerve agents. The enzyme active site contains two zinc atoms bound to the protein with His side chains and bridged by a carbamoylated Lys sidechain. The enzyme is remarkably efficient; the catalytic rate of phosphotriesterase is limited only by the diffusion rate of enzyme and substrate in solution. Because the phosphotriester substrates for this enzyme are synthetic compounds with no known biological counterpart, it is interesting to consider the evolutionary pathway that led to the invention of PTE--what was the starting point and how was this new activity created? Using homology searching techniques, we have found an E. coli genomic open reading frame (ORF) that is highly similar to phosphotriesterase. This ORF is located at about 74 min. on the E. coli chromosome in a cluster of newly identified ORF's that appear to encode enzymes used in organophosphate metabolism. We have named the E. coli homolog phosphotriesterase homology protein (PHP). The polypeptide sequence of PHP is 28% identical to PTE and the four His residues that are ligands to the Zn atoms in PTE are conserved in PHP. In addition, PHP contains a Glu residue at the corresponding position of the carbamoylated Lys of PTE. Based on this structural homoloy, we predict that: 1) PHP has an a/b barrel structure; 2) PHP is an enzyme; 3) PHP is a metallo enzyme; and 4) PHP is a member of the enzyme subfamily from which PTE originated. The relationship between PTE and PHP presents a unique research opportunity to study a pathway in enzyme evolution. PHP represents the start point and PTE the end point, and the question is whether we can recreate the mutations in the laboratory that convert PHP into PTE. We are also interested in understanding the natural biological function of PHP, as this may provide clues to why a PHP subfamily member was recruited by soil bacteria to become a new phosphotriester hydrolase.